The ever-increasing demand for smaller, faster and less expensive microelectronic and micro-electromechanical systems requires corresponding development of efficient and suitable manufacturing techniques.
Either additive or subtractive techniques are used in the fabrication of micro-, and/or nano-structures on a surface. One general subtractive technique is etching and one general additive technique is plating.
The etching methods are usually divided into two subgroups, dry- and wet etching. In general, dry etching is used for submicron structures and/or where straight sidewalls are important. Wet etching is used for large structures where some undercutting is acceptable or sometimes desirable. The wet etching techniques can be divided into chemical- and electrochemical etching.
The advantage of dry etching compared to wet etching is that anisotropic etched profiles can be generated in both crystalline and polycrystalline/amorphous material. Some of the disadvantages of dry etching are high equipment costs, lack of selectivity, problems with redeposition on the sample, environmentally hazardous chemicals, surface damages on the etched sample and safety and disposal problems.
The advantage of wet etching is that it is a simple and inexpensive process. One of the disadvantages is that it does not involve any directional driving force and therefore the etching rate is the same in all directions, which results in an isotropic etchprofile. Some other disadvantages are that wet etching baths generally contain aggressive and toxic chemicals, which results in safety and disposal problems. In many wet etching processes waste treatment and disposal costs often surpass actual etching costs, and the same drawback applies for dry etching.
Detailed descriptions regarding the above mentioned etching processes are considered known by a man skilled in the art and will not be presented in this paper. Because of the close relationship between the etching method according to the present invention and the electrochemical etching some details regarding the later will be presented as follows.
Electrochemical etching is a simple and inexpensive etching method, which makes it possible to achieve high etch rates and accurate process control. In electrochemical etching an external electrical potential is applied between an etched sample and a counter electrode, all immersed in a liquid etchant. An electrochemical cell with the working electrode, the sample, as anode and the counter electrode as cathode are formed, as shown in FIG. 1. An external potential is applied to drive the oxidation process at the working electrode. The corresponding reduction at the cathode is usually hydrogen gas formation. As electrolyte, and etchant, neutral salt solutions or very diluted mixtures of conventional etchant can be used. The applied potential and the electric field from it give a directional etching in the vertical direction.
One problem the designers of electrochemical etching cells are facing is that, to reduce the resistive losses from charge transfer in the electrolyte, one wants a small electrode distance. A small distance, which makes just a tiny unevenness in the electrode, give rise to a relatively big Δd that, gives a non-uniform current density distribution. The result is that some parts of the sample are over-etched while some parts are not etched to the desired depth. No mechanical support is possible to keep the electrode in position over the whole surface, since no contact between sample and counter electrode is allowed.
Another problem in electrochemical etching is non-uniform current density distribution arising from accumulated currents from non-etched areas, due to the fact that all parts of the counter electrode are in contact with the electrolyte, and not only the desired areas, above the etched parts.
The second option, additive techniques, for pattern transfer is to add material in the structure formed on top of the substrate by the pattern-defining step. Electrochemical deposition, for which the persons skilled in the art also use the term “electroplating”, physical vapour deposition and chemical vapour deposition are examples of additive processes. It is known in the field that, by using electroplating, well defined patterns, vertical sidewalls and high aspect ratio structures can be fabricated. However, common industrial problems are associated with the known electroplating process, namely non uniform current density distribution resulting in a deposition rate depending on the pattern surrounding each structure that is plated. Furthermore, such differences in current density also result in different material composition when plating alloys, as well as differences in height of electroplated structures on a substrate. Up to now, these undesired uneven distributions typically have to be rectified using planarization methods in a subsequent process step.
When the purpose of etching is to provide a structure in the etching material by etching away selected parts, thereof, the etching material which is not to be etched away is usually coated with an etching preventing layer, a so called mask or resist. The primary technique to define patterns to be etched is photolithography and a common etching preventing layer is a photo-resist. The photo-resist is exposed by electromagnetic radiation and developed to transfer the pattern where etching is wanted. Every sample that is etched has to be coated with resist, pre-baked, exposed, developed and hard-baked before the etching process can start.
Most of today's micro-devices are built up by a large number of functional layers and each layer has to be patterned and aligned in a photolithography process followed by a pattern transfer process. FIG. 6 shows a conventional etching process with the lithography process. The complicated nature of the pattern defining lithography process and the large number of lithography steps needed to fabricate a micro-device makes it to a major time and cost carrier in the total manufacturing chain.
From the European patent publication EP 1060299 it is known to use a method of making, by etching, depressions in selected portions of an etching surface by using an electrode with electrically conductive electrode portions in selected portions of an electrode surface, where the electrode portions is forming an electrode pattern which corresponds to the etching pattern. The method is different compared to the present invention by using electromagnetic radiation to dissolve a passivating layer, which is formed on the etching material. During etching the electrode is placed at a distance from the electrically conductive etching material, which also differs from the present invention. The electrodes according to EP 1060299 have to be transparent to electromagnetic radiation and they do not compensate for unevenness in the micro/nano areas.
WO 9845504 discloses a method for electroplating using an electroplating article, an anode and a substrate. The electroplating article is put in contact with the substrate. In one embodiment, the external anode is placed separated from the substrate and the electroplating article, all immersed in an electrolyte. According to the disclosure, a potential is applied over the external anode and the substrate, resulting in material transferred from the anode, through the porous carrier of the electroplating article and plated on the substrate in a pattern defined by the insulating mask of the electroplating article. The electrolyte volume between the electroplating article and the anode can be agitated to improve mass transfer of electroactive ions. However, the disclosed method struggles with the same problems and drawbacks as associated with conventional electroplating, namely non-uniform plating rates as a result of non-uniform current density distribution due to the anode having areas with a surface size differing from the surface size of corresponding cathode areas on the patterned substrate. Thus, differences in reaction rates in different cavities result in plated microstructures with different heights depending on the pattern surrounding each structure. The problem is usually solved by a subsequent planarization process step like lapping or CMP (Chemical Mechanical Polishing). When plating alloys, the method described in WO9845504 suffers from the same problems as conventional plating processes, namely differences in material composition because of non uniform current density distribution.
Furthermore, the mentioned embodiment disclosed in WO 9845504 requires an electroplating article fabricated with a porous material that is permeable for ions in the electrolyte, which gives rise to limitations in how small dimensions that can be defined, depending on the pore size of the material.
In a second embodiment disclosed in WO 9845504 it is mentioned an electroplating article that consists of a patterned mask placed onto an anode. The anode can be soluble or insoluble and can include an erodable layer. In the method using a soluble anode, the material is transferred from the anode material in the electroplating article, thus the electroplating article is eroded during use, but can be periodically redressed and reused. However, the problem of non uniform current density distribution also applies to this method, as the patterned mask still is placed as a separate layer onto the anode layer, i.e. the current density distribution is only at the beginning of a plating process uniform, whereas the contact surface of the electrolyte with the anode material increases differently in each local plating cell, depending on its size, as anode material is consumed. Moreover, the maximum aspect ratio, i.e. height/width ratio, of structures that can be plated is limited by the fact that the erosion of the anode material in the electroplating article undercuts the insulating pattern mask. Undercutting the mask layer during use is also associated with reliability problems, since the patterned mask layer will be completely undercut and disintegrated from the electroplating article if the electroplating process is not terminated in time. The problems described are inherently associated with the method because the soluble anodic material is transferred directly from the electroplating article itself, even in the case where the electroplating article consists of different layers of soluble and insoluble material.